On Thu, Jun 13, 2024 at 04:50:48PM -0700, Sean Christopherson wrote:
> On Thu, Jun 13, 2024, Dan Carpenter wrote:
> > This error path used to return -ENOMEM from the where r is initialized
> > at the top of the function. But a new "r = kvm_init_irq_routing(kvm);"
> > was introduced in the middle of the function so now the error code is
> > not set and it eventually leads to a NULL dereference. Set the error
> > code back to -ENOMEM.
> >
> > Fixes: fbe4a7e881d4 ("KVM: Setup empty IRQ routing when creating a VM")
> > Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> > ---
> > virt/kvm/kvm_main.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> > index 07ec9b67a202..ea7e32d722c9 100644
> > --- a/virt/kvm/kvm_main.c
> > +++ b/virt/kvm/kvm_main.c
> > @@ -1212,8 +1212,10 @@ static struct kvm *kvm_create_vm(unsigned long type, const char *fdname)
> > for (i = 0; i < KVM_NR_BUSES; i++) {
> > rcu_assign_pointer(kvm->buses[i],
> > kzalloc(sizeof(struct kvm_io_bus), GFP_KERNEL_ACCOUNT));
> > - if (!kvm->buses[i])
> > + if (!kvm->buses[i]) {
> > + r = -ENOMEM;
> > goto out_err_no_arch_destroy_vm;
> > + }
> > }
>
> Drat. Any objection to tweaking this slightly to guard against similar bugs in
> the future? If not, I'll apply+push the below.
No objections from me. :)
regards,
dan carpenter
